This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mhealth and uhealth, is properly cited. The complete bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license information must be included.
Children are 5 times more likely to be overweight at the age of 12 years if they are overweight during the preschool period.
The purpose of this study was to establish the feasibility, acceptability, and preliminary effects of a cognitive behavioral intervention (TEXT2COPE) synergized with tailored mobile technology (mHealth) on the healthy lifestyle behaviors of parents of overweight and obese preschoolers delivered in a primary care setting.
Fifteen preschooler-parent dyads recruited through primary care clinics completed a manualized 7-week cognitive behavioral skills building intervention. Beck’s Cognitive Theory guided the TEXT2COPE intervention content and Fogg’s Behavior Model guided the implementation. The intervention employed a combination of face-to-face clinic visits and ecological momentary interventions using text messaging (short message service, SMS). To enhance the intervention’s relevance to the family’s needs, parents dictated the wording of the text messages and also were able to adapt the frequency and timing of delivery throughout program implementation.
Self-reported findings indicate that the program is feasible and acceptable in this population. The intervention showed preliminary effects with significant improvements on parental knowledge about nutrition (
Utilizing a cognitive behavioral skills intervention with SMS has great potential for supporting clinical care of overweight and obese preschool children and their families. Further exploration of the potential effects on health and behavioral outcomes is warranted.
The preschool age group is a priority for surveillance and intense intervention because obesity trends in this group are a predictor of trends in older children and adults [
Evidence suggests that parents desire personalized information relevant to their child and are enthusiastic about receiving text messages (short message service, SMS) endorsed by their child’s primary care provider [
Therefore, the purpose of this study was to establish the feasibility, acceptability, and preliminary effects of a 7-session cognitive behavioral intervention combined with tailored and adaptive SMS regarding healthy lifestyle beliefs, perceived difficulty, and behaviors of parents of overweight and obese (OW/OB) preschoolers (aged 3-5 years) delivered in a primary care setting. The first aim was to examine the feasibility and acceptability of the intervention as determined by rates of participant retention, adherence to the study protocol, and response to participant evaluation. The second aim was to evaluate preliminary effects of the intervention on parental self-reported nutrition/physical activity knowledge, healthy lifestyle beliefs, perceived difficulty, and healthy lifestyle behaviors determined by pretest-posttest scores and effect sizes.
Previous evidence [
TEXT2COPE Intervention Program |
Session Mode |
Healthy lifestyles and the thinking, feeling, behaving triangle; basic recommendations for nutrition & physical activity in preschoolers; goal setting | Session 1 |
Information on physical activity and nutrition, including appropriate portion sizes, healthy eating, and food groups | Session 2 Self-study AND text messaging |
Barriers to goal progression and overcoming barriers through problem solving and cue recognition | Session 3 Face-to-face AND text messaging |
Positive thinking and self-talk related to healthy lifestyle behaviors | Session 4 Self-study AND text messaging |
Cognitive reframing—with an emphasis on physical and emotional responses to stress and how positive beliefs can help to reframe cognitions and promote positive coping | Session 5 Face-to-face AND text messaging |
Effective communication, stress, and coping | Session 6 Self-study AND text messaging |
Aspects of Fogg’s Behavior Model (FBM) informed the overarching strategy for the study [
Conceptual model of the TEXT2COPE program.
This pilot study used a one-group pre and posttest preexperimental design approved by the Arizona State University Institutional Review Board. Participants were recruited from 3 pediatric primary care practices. To be included, the primary caretaker (hereafter referred to as the
The principal investigator (PI) partnered with MEMOTEXT Corporation to develop software specific to this project for delivering/receiving SMS data. Participant confidentiality was secured through MEMOTEXT’s privacy policy and data were used in accordance with all applicable laws and the Personal Information Protection and Electronic Documents Act. Per the study protocol, no personal health information was communicated via SMS. Both static and tailored messages were entered into the SMS application used for this project. A sample of the software shown in
TEXT2COPE-Memotext application
The clinics preferred a combination of burst and serial recruitment efforts. In all of the offices, a support staff member (eg, medical assistant, receptionist) was identified to serve as a recruitment facilitator. The clinic staff member phoned and/or mailed a scripted recruitment letter to all participants identified as having a diagnosis of OW/OB based on provider referral or billing code, and who met eligibility criteria. Recruitment occurred from March to November 2013.
Flow diagram of participant recruitment.
The intervention implementation consisted of 4 key strategies: (a) face-to-face visits (covering educational information and cognitive behavioral skills), (b) reminders, (c) triggers, and (d) reinforcements.
TEXT2COPE intervention implementation.
Session | TEXT2COPE Intervention Program Cognitive Behavioral Skills Content | Mode |
1 | Healthy lifestyles and the thinking, feeling, behaving triangle; basic recommendations for nutrition & physical activity in preschoolers; goal setting | Face-to-face AND text messaging |
2 | Information on physical activity and nutrition, including appropriate portion sizes, healthy eating, and food groups | Self-study AND text messaging |
3 | Barriers to goal progression and overcoming barriers through problem solving and cue recognition | Face-to-face AND text messaging |
4 | Positive thinking and self-talk related to healthy lifestyle behaviors | Self-study AND text messaging |
5 | Cognitive reframing—with an emphasis on physical and emotional responses to stress and how positive beliefs can help to reframe cognitions and promote positive coping | Face-to-face AND text messaging |
6 | Effective communication, stress, and coping | Self-study AND text messaging |
7 | Putting it all together; integration of knowledge and skills | Face-to-face |
Participants responded to static test messages sent twice weekly. For simplicity, because of related health benefits, and evidence indicating that most preschoolers do not eat the recommended daily serving of vegetables, the authors collectively decided to use vegetable consumption as the topic of the static text messages. The use of 2-way weekly text messages (sent/received) is consistent with previous literature [
To enhance SMS support, parents were taught the difference between hot and cold triggers. At the end of each face-to-face visit, each parent developed a custom text message intended to hot trigger a skill at home. By having participants select the skill to work on, the general verbiage of the prompt, and the day(s) and time each week that they would like to receive the prompt, parents were able to reflect on their own family needs and tailor the intervention content accordingly. The goal was to customize the text message to the families and deliver it at just the right time. Parents were encouraged to create a text message that would “speak to them” regarding that week’s skill or goal. If a parent was unable to articulate a text message, the PI reviewed the session content with the participant until the parent was able to verbalize comprehension and subsequent SMS content. Test messages varied depending on the participant and included the following: the parents’ own words, a mixture of PI and parent words (up to 140 characters), examples from the manual tailored by participants, and a combination of a tip with a page reference from the manual for further information. Examples of the text messages crafted by parents included, “Run, laugh, chase the little guy around. Keep moving!!! Repeat it, BELIEVE IT! You’ve got this!!!!” or “Kick your own butt for 2 minutes! Follow your exercise plan and believe in yourself, now go!” “Wake up sleepy head - you'll regret it if you don't. RUN RUN RUN! you're the Role Model. Don't hit snooze!”
Since the individual ultimately chose the SMS content, each text message took on the tone preferred by the participant (ie, funny or authoritative).
Static text message and support.
Homework from the previous session was reviewed at each face-to-face visit. To ensure reception and comprehension, parents were asked to respond to the static text messages. An additional log for each text message sent and received was maintained throughout the study. Two of the authors reviewed all text messages.
A subset of the SMS data reflected a quality control population. Outside of the iterative testing phase of the MEMOTEXT-TEXT2COPE software, the PI included a confederate participant to receive text messages at the same time the MEMOTEXT system was
Measures used in this pilot study assessed parental demographics and child anthropometric measures, parental nutrition and physical activity knowledge as it related to their child [
Summary of data collection.
Construct | Instrument | Validity | Reliabilitya | Data Collection |
Demographics | Demographic Questionnaire |
|
NA | T1b |
Parent Evaluation | Exit Interview |
|
NA | T2c |
Healthy Lifestyles Beliefs Scale | Healthy Lifestyles Beliefs Scale | Content, Construct | .86 | T1, T2 |
Perceived Difficulty | Perceived Difficulty Scale | Content, Construct | .92 | T1, T2 |
Knowledge | Nutrition Knowledge | Content, Construct | .74 | T1, T2 |
Knowledge | Physical Activity Knowledge | Content, Construct | .52 | T1, T2 |
Healthy Lifestyle Behaviors - Parent | Healthy Lifestyle Behaviors Scale | Content, Construct | .78 | T1, T2 |
a Cronbach’s alpha
b T1: Time 1
c T2: Time 2
All analyses were conducted using SPSS 21. For normally distributed data (nutrition knowledge, perceived difficulty, healthy lifestyle behaviors), paired sample
Mothers were predominant study participants and caregivers, consistent with other family-based interventions [
Descriptive characteristics of TEXT2COPE parents and preschoolers.
Participant | Characteristic |
|
n (%) |
|
|
18-24 | 3 (24) |
|
|
25-34 | 5 (38) |
|
|
35-44 | 5 (38) |
|
|
Female | 13 (100) |
|
|
Male | 0 (0) |
|
|
Caucasian | 4 (31) |
|
|
African American | 1 (8) |
|
|
Hispanic/Latino | 8 (61) |
|
|
Less than high school | 1 (8) |
|
|
General equivalency diploma/high school | 5 (38) |
|
|
Some college | 3 (23) |
|
|
4-year college | 3 (23) |
|
|
Master’s degree | 1 (8) |
|
|
Less than $10,000 | 5 (38) |
|
|
$10,000-$19,000 | 4 (31) |
|
|
$20,000-$29,000 | 1 (8) |
|
|
$40,000-$49,000 | 1 (8) |
|
|
$70,000 or more | 2 (15) |
|
|
Yes | 11 (85) |
|
|
No | 2 (15) |
|
|
3 years | 2 (13) |
|
|
4 years | 5 (33) |
|
|
5 years | 8 (54) |
|
|
Female | 5 (33) |
|
|
Male | 10 (67) |
Parents reported the program to be helpful 100% of the time and 100% of parents indicated that they would recommend TEXT2COPE to other parents of preschoolers. Qualitative feedback included, “The program doesn’t just tell you to be healthy, it shows you how,” and participants noted that it included messages to “Keep trying, don’t give up,” and “Keep pushing towards my goals.” Thematic saturation occurred for goal setting (10/13, 77%) as the most-liked content of the program. No saturation occurred for topics least liked.
Paired
Self-reported cognitive and behavioral outcomes for the TEXT2COPE group.
Instrument | Pretest Mean (SD) | Posttest Mean (SD) | Effect Size |
|
Nutrition Knowledge | 14.60 (3.46) | 17.53 (1.73) | 1.07a | .001 |
Activity Knowledge | 9.93 (1.58) | 11.00 (1.07) | .45b | .012 |
Parental Healthy Beliefs | 79.53 (8.10) | 83.00 (7.63) | .61b | .001 |
Parental Healthy Behaviors | 46.20 (8.37) | 51.00 (7.83) | .59a | .040 |
Perceived Difficulty c | 38.46 (11.33) a | 42.31 (10.94) a | .50a | .168 |
a Effect size for Cohen’s
b Effect size for R code (nonparametric): .1 small, .3 medium, .5 large
c Higher scores on the Perceived Difficulty scale indicate less perceived difficulty (easier to do)
Parent participants were defined as completing the program if all of the content in the 7 TEXT2COPE sessions were covered and parents completed pretest and posttest measures. Findings reflect a 100% retention rate. Although missed or canceled appointments did occur, participants were able to continue the program through provider understanding, rescheduling of appointments, and counseling on the benefits of healthy lifestyle habit establishment relevant to the family’s needs. It is believed that customization of the content relative to the family, routine follow-up in person, and supplemental text messages facilitated patient engagement.
The 4 face-to-face visits, which lasted approximately 20-30 minutes to mimic the length of a standard office visit in primary care, were conducted during an outpatient visit at the child’s primary care office. As in other studies [
A total of 291 text messages were sent to parents: 138 tailored text messages, 105 static text messages, and 53 automated feedback text messages. The number of text messages sent to each participant varied from 7 to 39, with a mean of 22.31 (SD 9.47). Tailored text messages per participant ranged from 3 to 20 (mean 10.62, SD 5.45). The number of static text messages sent to each participant ranged from 4 to 12 (mean 8.00, SD 2.42). Overall, 69% of parents (9/13) changed the frequency of tailored text messages, increasing delivery from once weekly to as many as 5 times a week. Timing (when the message was delivered) was evaluated by dividing the day into 3 segments: morning (6:01 am-12:00 pm), afternoon (12:01 pm-6:00 pm), and evening (6:01 pm and later). Parents favored receiving tailored text messages in the afternoon (58.8% of the time), coinciding with parents’ perceived need for support (and when the message would most likely serve as a hot trigger). Peak timing occurred during the hours when children were home from school, the workday ended, and dinner/afterschool activities coincided. All parents reported SMS frequency to be “just right,” while 85% felt the timing of delivery was “perfect” (n=11/13).
None of the tailored text messages were used strictly as a reminder. All of the tailored text messages incorporated an aspect of the cognitive behavior skills associated with that week’s session. For example, parents were encouraged to make a plan and set SMART (specific, measurable, attainable, realistic, and timely) goals. After learning about that skill, a participant used the SMS to trigger physical activity with the message, “[Parent name]- Make a plan; schedule a SMART goal for family fun exercise; Remember- set yourself UP for SUCCESS! Believe you can do anything!” This message was delivered twice in one week, during the afternoon hours when the parent could make a plan and act on it. As parents advanced through the program, it appeared that they became more aware of how SMS content, timing, and frequency could cue them to act (or highlight an inability to act). It also became apparent that participants liked having a reminder of content covered in the face-to-face visit and corresponding section of the manual/audio CD to help them work on skills. For instance, one participant reported having the television on when her son was home in the afternoon, but wanted to work on increasing physical activity. She developed a combination text message with a reminder of where in the manual she could find help (Chapters 5 and 6), but also a tip on how she could build in a skill to meet her goal. At 4:30 pm, on a specified day, her text message was sent stating, “You're near the finish line. This wk 5&6! Decrease TV time or family exercise challenge on commercials! Who will win?”
Parents reported not responding to the static text messages because they “thought it was a reminder” or “didn’t know to respond back.” The initial response rate was only 26% (15/58) for the static text messages. Subsequently, verbiage of the static text messages was made shorter and simpler, while topic and response algorithms remained unchanged. The response rate improved to 80% (38/47) with this change, ranging from 0%-100% response per participant (mean 49%, SD 36.2%). Overall, 87% of parents reported offering their child vegetables at dinner. Messages were confirmed delivered. The sociodemographic variability of the small sample was a factor in two-way SMS. At least two families had very limited SMS data plans. As a result, they could continue to receive text messages, but could not send them. Therefore, one participant did not respond to any of the static text messages (yes/no questions). Another parent circumvented her data plan by using a free software application to send outgoing text messages. While the service was free, it altered every outgoing message with a tag line advertising their service (eg, “Y -Sent free from TextNow.com”). This tag line was not part of the TEXT2COPE if/then algorithm, and subsequently fell into an aberrant response category. As a result, that participant did not receive immediate automated feedback to her response.
The results of this study indicate that a cognitive behavioral skills program, synergized with mobile messaging, is feasible and acceptable in a primary care setting with parents of OW/OB preschool-aged children. The preliminary short-term effects of this research show great potential for promoting healthy lifestyle choices in this population. There are a number of key points to be learned from this research. Using recommendations from Bowen [
Parents reported high satisfaction with the TEXT2COPE program and expressed intent to continue using the skills learned in the program. The skills covered in the intervention align with previous research and perhaps shed light on retention and acceptance rates as well as preliminary effects. Better parental compliance with behavior strategies (eg, targeting specific behaviors, self-monitoring, goal setting, stimulus control/environmental cues, positive parenting strategies) is predictive of better outcomes for the child [
The use of static text messages warrants further exploration. The static text messages performed a multitude of functions. First, they served as a data collection tool with a varied response rate. The automated response built from the if/then algorithm served as reinforcement. Some parents viewed the text messages as reminders and did not respond. It is difficult to assess if response rate was a function of (a) the change in verbiage, (b) the way it was presented, or (c) characteristics of the participants responding. Predetermined delivery timing and vernacular could have contributed to the variability in response.
Demand for innovative programs in the primary care setting reflects two domains: demand from the organization and demand from the patients. Office managers, providers, and staff welcomed the TEXT2COPE program as an innovative resource to treat OW/OB. However, partnering with primary care clinics was difficult. Administrative affiliations warranted lengthy negotiations between the PI and the organizations and required academic institutional review board approval, legal review, and/or third party reviewers. Parents stated that they welcomed the SMS communication from their providers. However, outside of the TEXT2COPE program, we were unaware of any direct communication between clinic and patients, outside of traditional phone calls or appointments. One clinic had an established Facebook page, yet, the clinic was unaware of how many families subscribed to their page or received updates. Parents also voiced an interest in learning more about how to help their child, but often lacked the resources. Parents reported feeling more proactive in their family’s health while using clinic time to address skills that could promote behavior change. Supplementing these new skills with text messaging provided additional support for parents in their home environment.
The intervention was manualized to facilitate reproducibility. A health care provider or lifestyle coach could deliver the content. With regards to SMS, sending weekly text messages took less than 5 minutes per participant, per text message. Automating text message delivery using machine learning techniques or learned user preferences is possible with advances in mHealth. From a logistical perspective, the use of text messages may require additional software, which would be either independent or programmed into existing electronic health record (EHR) software.
Practicality was explored when resources, existing means, and time were constrained in some way. Given the nonacute nature of healthy lifestyle behaviors, parents often requested to reschedule appointments. Parents often reported that some habits were more difficult to change than others. Mothers commonly reported that positive self-talk and “believing is achieving” was not something they commonly practiced; however, after reviewing this content more, mothers appreciated how this could affect the way serve as a role model to their children. Equally as impressive was that mothers welcomed content about topics such as overcoming barriers, problem solving, and goal setting. Positive reactions indicated parents favored an emphasis on “
The user determined the number of weeks needed for content delivery, timing, and frequency of text message delivery. Most reflected parents’ routines around primary care use and their daily demands. However, adapting the schedule, content, and utilization of parent-driven text messages enhanced relevance to the user. Taveras and colleagues [
In terms of perceived fit, perceived sustainability, and costs to organizations and policies [
This research sheds light on programs aimed at families with OW/OB preschoolers in a primary care setting, which have been sparse in the literature to date. Using a dynamic (versus static) design, the research supported participants as they progressed toward their goals. Parents want specific, action-oriented advice to achieve goals rather than general information on healthy behaviors [
Although novel in approach, the findings must be tempered due to some limitations. First, the lack of an attention control group threatened the internal validity. Secondly, although fathers were encouraged to attend, the small sample size of only mothers (roughly 47% reported being single) limits the generalizability of the study findings. Perhaps a larger sample size and varying recruitment methods targeting dad’s groups could help to strengthen participation. Thirdly, measures were self-reported, wherein bias or human error can be introduced. Parent anthropometric measures would have been useful to obtain, as evidence suggests that one of the strongest predictors of childhood obesity is parental obesity [
Mobile technologies, in conjunction with clinical care, have the potential to expand care and reduce costs. This new delivery method is poised to reinvent health care delivery, as it is superior to other technologies due to high demographic penetration, an omnipresent nature, fluidity of use, and a broad range of capabilities [
Future research should involve a full-scale randomized controlled trial to determine the short- and long-term efficacy of this intervention with families who have OW/OB preschool children. Technology also lends itself to alternate research methods such as adaptive interventions. Highly adaptive sequential multiple assignment randomized trials (SMART) have been used to inform the best sequencing of treatments when individuals are not responding, and the best time to transition from more to less intensive or maintenance therapy [
cognitive behavioral therapy
Cognitive Theory
Fogg’s Behavior Model
overweight/obese
short message service
The authors would like to thank all the participants and clinic staff for their valuable time and input. The authors also wanted to give special thanks to Dr Gance-Cleveland and Dr Fenbert for their clinical expertise and support. Elisha, Jeff, and Amos are sincerely appreciated for their dedication and patience while developing the application used for this project.
Bernadette Melnyk is the founder of COPE 2 Thrive, an LLC that distributes the COPE intervention.